JANUARY 2015 (Revised) A PRINCIPLED TECHNOLOGIES TEST REPORT Commissioned by Dell, Inc. ORACLE RAC PERFORMANCE: DELL POWEREDGE FX2S WITH FLUID CACHE FOR SAN VS. COMPETING CACHE SOLUTION When upgrading database servers, an essential IT strategy for many enterprises, businesses are looking for powerful hardware that offers caching and flash-based solutions to help maximize performance. Adding flash technology directly into servers can be an attractive, high-performance option for both caching and storing data, but the number of required drives for sufficient capacity and fault tolerance for each individual workload may be cost-prohibitive. Dell Fluid Cache for SAN offers an attractive alternative. This innovative server-side read/write caching solution leverages high-speed NVMe PCIe SSDs in validated servers and low-latency private cache network technology to form a cache pool. When connected to a Dell Compellent SAN, this solution provides a highly available data caching solution that can dramatically improve application performance. The new Dell PowerEdge FX2s chassis populated with PowerEdge FC630 servers packs significant compute power into only 2U of rack space. When equipped with Dell Fluid Cache for SAN, this solution provides outstanding OLTP database performance. In our hands-on tests in the Principled Technologies labs, we compared the OLTP database performance of this solution to a legacy solution using three-year-old HP ProLiant DL380 Gen8 servers with a competing caching solution. We configured both Dell Fluid Cache for SAN and the legacy servers with competing cache solution with an Oracle Real Application Cluster (RAC). For the Dell Fluid Cache for SAN solution, we first configured it with two nodes to compare to the legacy servers with competing cache solution, then with four nodes to demonstrate the peak performance capabilities of four
35
Embed
Oracle RAC performance: Dell PowerEdge FX2 with Fluid ... · A PRINCIPLED TECHNOLOGIES TEST REPORT Commissioned by Dell, Inc. ORACLE RAC PERFORMANCE: DELL POWEREDGE FX2S WITH FLUID
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
JANUARY 2015 (Revised)
A PRINCIPLED TECHNOLOGIES TEST REPORT Commissioned by Dell, Inc.
ORACLE RAC PERFORMANCE: DELL POWEREDGE FX2S WITH FLUID CACHE FOR SAN VS. COMPETING CACHE SOLUTION
When upgrading database servers, an essential IT strategy for many enterprises,
businesses are looking for powerful hardware that offers caching and flash-based
solutions to help maximize performance. Adding flash technology directly into servers
can be an attractive, high-performance option for both caching and storing data, but the
number of required drives for sufficient capacity and fault tolerance for each individual
workload may be cost-prohibitive. Dell Fluid Cache for SAN offers an attractive
alternative. This innovative server-side read/write caching solution leverages high-speed
NVMe PCIe SSDs in validated servers and low-latency private cache network technology
to form a cache pool. When connected to a Dell Compellent SAN, this solution provides
a highly available data caching solution that can dramatically improve application
performance.
The new Dell PowerEdge FX2s chassis populated with PowerEdge FC630 servers
packs significant compute power into only 2U of rack space. When equipped with Dell
Fluid Cache for SAN, this solution provides outstanding OLTP database performance.
In our hands-on tests in the Principled Technologies labs, we compared the
OLTP database performance of this solution to a legacy solution using three-year-old HP
ProLiant DL380 Gen8 servers with a competing caching solution. We configured both
Dell Fluid Cache for SAN and the legacy servers with competing cache solution with an
Oracle Real Application Cluster (RAC). For the Dell Fluid Cache for SAN solution, we first
configured it with two nodes to compare to the legacy servers with competing cache
solution, then with four nodes to demonstrate the peak performance capabilities of four
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
FC630 servers running both RAC and Dell Fluid Cache for SAN. The Dell Fluid Cache for
SAN solution delivered 3.7 times the Oracle database transactions per minute of the
competing solution with two nodes, and 4 times the performance with four nodes,
making it an excellent choice over the legacy servers with competing cache solution.
WHAT IS DELL FLUID CACHE FOR SAN? Dell designed Fluid Cache for SAN to accelerate application IO, and boost
transactional performance for both read and write operations. Fluid Cache for SAN
utilizes Dell PowerEdge Express Flash NVMe PCIe SSDs in validated servers to create a
high performance cache pool. All cache pool servers each contain a Mellanox Connect X-
3 card and connect into a validated Dell Networking switch for the low-latency private
cache network. All of the servers are then connected to a Dell Compellent SAN.
The solution offers high availability by replicating blocks of data between cache
nodes connected to the high-speed RDMA network, so your data remains safe in cache
even in the unlikely event of a node or PCIe SSD failure in the cache pool. Fluid Cache for
SAN provides a single management interface and other capabilities such as cache-aware
snapshots, replication, and compression using Compellent Enterprise Manager. Dell
Fluid Cache for SAN also allows you room to grow, with the ability to scale up to eight
nodes per cache pool, scale up cache to 1.6 TB per server, and attach multiple cache
pools to one Compellent SAN to meet the future demands of your business.
Figure 1 shows an example environment highlighting how Fluid Cache for SAN
communicates between server and storage. Three validated nodes are required to run
Dell Fluid Cache for SAN software. Two of these three nodes are required to have a
minimum of one Dell PowerEdge Express Flash NVMe PCIe SSD each, and the network
for the private cache network is based on the low-latency RDMA protocol. The RDMA
network handles the cache pool communication between the nodes. Administrators use
the Dell Compellent Enterprise manager software to create and manage the Fluid Cache
cluster through automatic cache server discovery.
A Principled Technologies test report 3
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
Figure 1: Dell Fluid Cache for SAN communication.
OVERVIEW We first tested our database workload with two HP ProLiant DL380 Gen8
servers running Oracle 12c database Enterprise Edition in an Oracle RAC configuration
running on Red Hat Enterprise Linux (RHEL) 6.4. This solution used a competing cache
solution. We also configured a new Oracle RAC environment on the Dell PowerEdge
FX2s and FC630 servers upgraded to RHEL 6.5, and added Fluid Cache to the picture to
test the performance improvements. Testing the Dell solution, we first configured the
four Dell PowerEdge FC630 servers with two nodes in an Oracle RAC with all four
running Fluid Cache for SAN. We compared this configuration to the legacy server with
competing cache solution. We then added the third and fourth cache nodes to our
Oracle RAC to demonstrate the overall performance of all FX2s nodes running the
database. (For system configuration information, see Appendix A and for step-by-step
test details, see Appendix B.)
A Principled Technologies test report 4
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
WHAT WE FOUND Figure 2 shows the normalized test results, as Oracle licensing prohibits
disclosure of the actual results. The Dell Fluid Cache for SAN solution provided 3.7 times
the number of database transactions per minute (TPM) in a two-node Oracle RAC and
four times the number of TPM in a four-node RAC that the competing cache solution
provided.
Figure 2: The Dell PowerEdge FX2s with Dell Fluid Cache for SAN achieved up to four times the transactions per minute of the two-node legacy HP server with the competing cache solution.
CONFIGURATION AND TEST DETAILS Competing cache solution
For the competing solution, we configured two HP ProLiant DL380 Gen8 servers
with Red Hat Enterprise Linux 6.4. We then configured a two-node Oracle 12c database
Standard Edition RAC cluster on the HP ProLiant DL380 Gen8 servers with a 460GB
database leveraging Oracle Automatic Storage Management. For the cluster
management and ASM traffic, we attached the first port of a dual-port 10Gb card in
each DL380 server to a Dell Networking S4810 switch. For the client traffic, we attached
one of each server’s onboard 1Gb NICs to a Dell Networking 6248 switch. We then
configured the servers with the competing cache solution, utilizing a PCIe-SSD card and
cache management software, as well as mapped the storage volumes on the Dell
Compellent SC8000 storage (see Figure 3.)
A Principled Technologies test report 5
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
Figure 3: The competing cache solution configuration we tested.
Dell Fluid Cache for SAN solution We configured two Dell PowerEdge FC630 servers in an Oracle RAC. We then
installed the Dell Fluid Cache software and configured a dual-port 40Gb NIC in each
FC630 for Fluid Cache traffic. The two ports were split, connecting to two different Dell
Networking S4810 switch for redundancy. To complete our Fluid Cache cluster and add
cache devices, we configured each of the four Dell PowerEdge FC630 servers with one
800GB Dell PowerEdge Express Flash NVMe PCIe SSD (see Figure 4). For our last
scenario, we configured Oracle RAC on the remaining two nodes and added them to the
existing cluster.
We then installed the Fluid Cache software and configured the management
and Fluid Cache networking on all four servers. We then configured a Fluid Cache cluster
via Dell Compellent Enterprise Manager. To enable Fluid Cache, we unmapped the
volumes on the Dell Compellent SC8000 from the HP ProLiant DL380 Gen8 servers and
competing cache solution, and remapped them to the Dell PowerEdge FC630 servers,
enabling Fluid Cache with default settings.
A Principled Technologies test report 6
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
Figure 4: The Dell Fluid Cache for SAN configuration we tested.
Storage For our storage, we configured two Dell Compellent SC8000 controllers
connected to two Compellent SC220 disk enclosures, filled with rotational HDDs (48 x
146GB 15K drives). We configured a separate server to run the Dell Compellent
Enterprise Manager Suite. For our storage LUNs, we configured two 500GB LUNs for
backups, one 21GB LUN for the cluster registry volume, four 200GB LUNs for ASM data,
and four 30GB LUNs for ASM logs. We connected each server to Compellent via Fibre
Channel using a Brocade 300 Fabric switch. We then mapped each volume to two Dell
FC630 servers.
A Principled Technologies test report 7
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
ABOUT THE DELL POWEREDGE FX2S ENCLOSURE The shared infrastructure approach of the Dell PowerEdge FX2s enclosure
increases flexibility and can help you make the most of your data center space while
reducing the cables and switches needed to run your servers. The Dell PowerEdge FX2s
enclosure has a standard 2U footprint and features a modular design that can hold
different combinations of compute and storage nodes to meet your specific goals. The
PowerEdge FX2s fits four half-width or eight quarter-width compute nodes to increase
the compute density in your rack and optimize the space in your data center. You can
deploy the FX2s solution like a traditional rack-mounted server while gaining the
benefits and features that more expensive dense blade solutions provide. Important
features of the FX2s enclosure include:
Up to eight low-profile PCIe expansion slots
Two pass-through or optional networking FN I/O Aggregator modules
Embedded network adapters within the server nodes
Offers both chassis-based management through the Chassis
Management Controller and rack-based management through
Integrated Dell Remote Access (iDRAC)with Lifecycle Controller on each
compute node
The Dell PowerEdge FX2s enclosure fits a number of server and storage options,
including the PowerEdge FC430, FC630, and FC830 servers, and PowerEdge FD332
storage – all Intel-powered. The Dell PowerEdge FX2s solution we tested included four
half-width Dell PowerEdge FC630 server sleds powered by Intel Xeon processors E5-
2650 v3.
For more information about the Intel-powered Dell PowerEdge FX2s solution,
visit www.dell.com/us/business/p/poweredge-fx/pd.
ABOUT OUR TESTING For our workload, we ran HammerDB v2.16 with a TPC-C-like workload. We ran
each test with 101 users for a 30-minute ramp-up time and a 60-minute test duration.
We ran three tests with no Fluid Cache and three tests with Fluid Cache enabled on the
ASM Data LUNs. The 3.7 times and 4.0 times advantages are reported from the median
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
} elseif {$choice <= 23} {
if { $KEYANDTHINK } { keytime 2 }
ostat $curn_os $w_id $RAISEERROR
if { $KEYANDTHINK } { thinktime 5 }
}
}
oraclose $curn_no
oraclose $curn_py
oraclose $curn_dl
oraclose $curn_sl
oraclose $curn_os
oralogoff $lda
}
}
A Principled Technologies test report 35
Oracle RAC performance: Dell PowerEdge FX2s with Fluid Cache for SAN vs. competing cache solution
ABOUT PRINCIPLED TECHNOLOGIES
Principled Technologies, Inc. 1007 Slater Road, Suite 300 Durham, NC, 27703 www.principledtechnologies.com
We provide industry-leading technology assessment and fact-based marketing services. We bring to every assignment extensive experience with and expertise in all aspects of technology testing and analysis, from researching new technologies, to developing new methodologies, to testing with existing and new tools. When the assessment is complete, we know how to present the results to a broad range of target audiences. We provide our clients with the materials they need, from market-focused data to use in their own collateral to custom sales aids, such as test reports, performance assessments, and white papers. Every document reflects the results of our trusted independent analysis. We provide customized services that focus on our clients’ individual requirements. Whether the technology involves hardware, software, Web sites, or services, we offer the experience, expertise, and tools to help our clients assess how it will fare against its competition, its performance, its market readiness, and its quality and reliability. Our founders, Mark L. Van Name and Bill Catchings, have worked together in technology assessment for over 20 years. As journalists, they published over a thousand articles on a wide array of technology subjects. They created and led the Ziff-Davis Benchmark Operation, which developed such industry-standard benchmarks as Ziff Davis Media’s Winstone and WebBench. They founded and led eTesting Labs, and after the acquisition of that company by Lionbridge Technologies were the head and CTO of VeriTest.
Principled Technologies is a registered trademark of Principled Technologies, Inc. All other product names are the trademarks of their respective owners.
Disclaimer of Warranties; Limitation of Liability: PRINCIPLED TECHNOLOGIES, INC. HAS MADE REASONABLE EFFORTS TO ENSURE THE ACCURACY AND VALIDITY OF ITS TESTING, HOWEVER, PRINCIPLED TECHNOLOGIES, INC. SPECIFICALLY DISCLAIMS ANY WARRANTY, EXPRESSED OR IMPLIED, RELATING TO THE TEST RESULTS AND ANALYSIS, THEIR ACCURACY, COMPLETENESS OR QUALITY, INCLUDING ANY IMPLIED WARRANTY OF FITNESS FOR ANY PARTICULAR PURPOSE. ALL PERSONS OR ENTITIES RELYING ON THE RESULTS OF ANY TESTING DO SO AT THEIR OWN RISK, AND AGREE THAT PRINCIPLED TECHNOLOGIES, INC., ITS EMPLOYEES AND ITS SUBCONTRACTORS SHALL HAVE NO LIABILITY WHATSOEVER FROM ANY CLAIM OF LOSS OR DAMAGE ON ACCOUNT OF ANY ALLEGED ERROR OR DEFECT IN ANY TESTING PROCEDURE OR RESULT. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC. BE LIABLE FOR INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES IN CONNECTION WITH ITS TESTING, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. IN NO EVENT SHALL PRINCIPLED TECHNOLOGIES, INC.’S LIABILITY, INCLUDING FOR DIRECT DAMAGES, EXCEED THE AMOUNTS PAID IN CONNECTION WITH PRINCIPLED TECHNOLOGIES, INC.’S TESTING. CUSTOMER’S SOLE AND EXCLUSIVE REMEDIES ARE AS SET FORTH HEREIN.